OMB Releases First Government-wide Policy the Use of AI in Federal Agencies

?On March 28, 2024, OMB released memo M-24-10, focusing on mitigating risks associated with AI use while allowing Federal agencies to leverage its capabilities to deliver services throughout the government and to citizens. The memo emphasized the need for “responsible AI development and deployment,” with specific actions directed towards enhancing AI safety and security, protecting privacy, advancing equity and civil rights, and promoting innovation.

Key elements of the memo include:

? Risks and Safeguards: Federal agencies are tasked with identifying and managing AI risks, particularly those that could impact individual rights or safety. By December 1, 2024, agencies must implement safeguards to assess, test, and monitor AI's impacts, mitigate algorithmic discrimination, and ensure public transparency in AI use.

? Transparency and Public Engagement: Agencies are required to develop strategies to ensure public transparency regarding their use of AI, including releasing annual inventories of AI use cases, reporting metrics for sensitive AI use cases, and notifying the public of any AI exempted from compliance. The memo also encourages the release of government-owned AI code, models, and data to facilitate public engagement and collaboration.

? Innovation and Workforce Development: The memo encourages agencies to leverage tools such as generative AI to help their workforce address challenges such as the climate crisis, public health, and public safety. It also directs agencies to expand and upskill their AI talent, highlighting initiatives like hiring 100 AI professionals and running a career fair for AI roles across the Federal Government.

? Governance and Accountability: Federal agencies are required to designate Chief AI Officers (CAIOs) and establish AI Governance Boards to oversee the use of AI and ensure accountability, leadership, and oversight in AI implementation.

For companies and contractors working with federal agencies, this guidance introduces new considerations, including compliance and reporting requirements, interoperability and usage rights, and the need to share code and data. If recent regulatory proposals are any indication, such as those released earlier this year regarding cybersecurity inventories and zero-trust, contractors can soon expect to see these Agency regulatory frameworks apply to their responsible use of AI throughout the entire Federal procurement ecosystem.

#AI #innovation #ResponsibleAI

M-24-10-Advancing-Governance-Innovation-and-Risk-Management-for-Agency-Use-of-Artificial-Intelligence.pdf (whitehouse.gov)